In observing an optically transparent micro-sample including a bio-sample like cells and tissues under a general optical microscope, there are some problems that it is very difficult to obtain a clear morphological image of various intracellular organs and materials in the sample and also it is impossible to selectively measure a spatial distribution of molecular chemical species. This is because a difference between objects and a background substance is minute in optical properties and thus a sufficient optical contrast therebetween is not provided. In other words, it is not easy to distinguish the specific micro-structures or micro-materials to be observed from the background substance in the sample.
To explore phenomena of life science and disease mechanisms by observing behavior of intracellular organs or metabolites in cells through optical images, a laser scanning fluorescent microscope which could overcome a limitation of a general optical microscope has been widely used for a long time. According to an operation principle of the laser scanning fluorescent microscope, the sample is dyed with a fluorescent marker which selectively combines with the objects to be observed, and then fluorescence generated by scanning ultraviolet or visible laser irradiation is detected spatially, thereby obtaining an optical image in which the objects gain high contrast selectively. However, since the fluorescent marker as an exogenous material is added in the bio-sample, there is a basic problem that an original status of the bio-sample is not maintained as it was. Because the added exogenous material lowers activity of the bio-sample, it is difficult to obtain exact information about the behavior of the bio material. Furthermore, since a coloring matter of the fluorescent marker is easily photobleached even by very weak lasers, it is also difficult to observe the image continuously or in time-lapse measurements.
In order to avoid the disadvantages of the laser scanning fluorescent microscope, there has been proposed a new microscopic technology which can detect a spectroscopic characteristic of a material itself without the fluorescent marker which is selectively combined with molecules of the material and thus obtain a molecular image. As a representative method of analyzing spectroscopic signals of its own molecular vibrational fingerprint generated by interaction between particular molecules and laser beams, there are infrared absorption spectroscopy and Raman scattering spectroscopy which are combined with a microscopic optical system so as to be used for molecular image measurement of a micro-structure.
FIG. 1 is a diagram of a molecular vibrational transition showing the principle of measuring an infrared absorption spectroscopic signal.
An infrared absorption microscope uses the principle that, when a laser beam scanned in a sample has a wavelength which resonates with inherent molecular vibration, the laser beam is strongly absorbed by the sample, and the intensity of the laser which is transmitted or reflected is reduced. The infrared absorption molecular image is obtained by measuring an attenuation ratio of the laser as a function of wavelength with respect to a scanning position of the laser beam.
FIG. 2 is a diagram of a molecular vibrational transition showing the principle of generating a spontaneous Raman spectroscopic signal.
In a Raman microscope, pixel data constructing a molecular image is formed by a spontaneous Raman spectrum generated by scanning the laser of a predetermined wavelength on the sample. Unlike in the infrared absorption microscope, a red-shifted Raman spectroscopic signal is generated by inelastic scattering of photons of the laser having a fixed wavelength which does not resonate with the molecular vibration of the objects. In this case, energy difference between incident photons and Raman-scattered photons corresponds to molecular vibration mode energy of materials in the sample. In other words, the Ramna spectrum obtained by collecting the laser beam scattered by the sample includes information about the inherent molecular vibration mode of the materials constituting the sample.
The infrared absorption microscope and the Raman microscope have various advantages and disadvantages, respectively.
Since the infrared absorption microscope is based on a linear absorption phenomenon, it is very simple to understand the principle and to analyze the signal. Furthermore, since the infrared absorption microscope uses a direct absorption spectroscopic signal generated from the resonance of the excitation laser and the molecular vibration mode, it offers high sensitivity and signal-to-noise ratio in the measurement. However, in the mid-IR regime with wavelength range of 2.5˜18 μm where light is resonated with the molecular vibration and thus strongly absorbed, it is in practice very difficult to realize a wavelength-tunable laser beam source which is required for the microscopic image. Furthermore, due to the limitation imposed by diffraction phenomenon, the infrared absorption microscope has a very low spatial resolution of 10˜40 μm compared with the spatial resolution (of 0.3˜0.5 μm) in an optical microscope with a visible beam source. To solve the problem of lacking tunable mid-IR sources, a wavelength dispersive detector or an interferometer for fourier-transform spectroscopy is employed in the measuring part, in order for a white beam source having a wide spectrum from the visible to the mid-IR to be used.
On the contrary, since the Raman microscope uses a single wavelength beam source not related to a molecular vibrational frequency, the excitation laser is given less weight in constructing the Raman microscope, and its operation is very simple. Furthermore, since the Raman microscope uses a laser source having a short wavelength of the visible light range, it has an advantage of obtaining the microscopic image having a good spatial resolution. However, it has also a disadvantage that it takes a long time to obtain an image because the intensity of Raman signal for providing spectroscopic information is very weak. Particularly, in the case that a dynamic characteristic of the living bio-sample is observed, or an intensity of excitation laser cannot be sufficiently increased to avoid damage to the sample, the disadvantage becomes further serious.
Research and development has been carried out continuously to overcome the above disadvantages of the infrared absorption microscope and the Raman microscope.
In order to improve the low spatial resolution of the infrared absorption microscope, there has been proposed a scanning near-field IR microscope using metal-coated optical fiber in which an aperture of a few hundred nm is formed at a distal end of a tip. Herein, the spatial resolution does not depend on the IR diffraction limitation, but depends on the size of aperture formed at the tip of the optical fiber, regardless of the wavelength of a laser beam, thereby providing the spatial resolution comparable with that in the optical microscope using a visible beam source. However, in order to obtain the microscopic image, it is necessary for a surface of the sample to be mechanically scanned with high precision relative to the tips of the optical fiber, but it is difficult to quickly obtain the image since the aperture has a very low transmission efficiency (˜10−6) at IR wavelengths. In addition, with a current technical standard, it is difficult to provide a reliable process technology which manufactures the optical fiber for efficiently transferring the IR. That is, it overcomes the limitation of the spatial resolution, whereas it submits to a sacrifice of the measurement sensitivity and the speed for obtaining the microscopic image.
A coherent andti-Stokes Raman scattering (CARS) microscope overcomes substantially low sensitivity and slow speed for obtaining the microscopic image which were critical disadvantages in the conventional Raman microscope. In detecting of molecular vibrations, CARS is also based on the Raman scattering phenomenon is similar to that in the conventional Raman microscope. However, the fundamental difference is not to use the spontaneous Raman scattering as a linear optical phenomenon, but to use a four wave mixing in which three incident laser beams are interacting with the sample so as to generate a nonlinear optical signal.
FIG. 3 is a diagram of a molecular vibrational shift showing a principle of generating a CARS nonlinear spectroscopic signal.
Referring to FIG. 3, two incident laser beams (pump beam and Stokes beam) having a frequency difference corresponding to Raman shift of a certain molecule in the sample generate a beat and then induces forced harmonic molecular vibration which is coherent with the beat waveform. If a third laser beam (probe beam) is incident on the molecules which are vibrating in phase with coherence a status that phases are harmonized, anti-Stokes Raman scattering whose wavelength becomes shorter takes place through interaction, resulting in a coherent signal beam having the same phase in a predetermined propagating direction. Then, the nonlinear optical signal is precisely mapped in a space of the sample, thereby obtaining the CARS microscopic image.
The CARS microscope has an advantage of providing a very high measurement sensitivity and high speed for obtaining the image as well as obtaining the selective image. Since the CARS generates a very intense signal beam than the spontaneous Raman scattering, it is possible to quickly obtain high quality images having a good signal-to-noise ratio. The CARS phenomenon depends on the characteristic of third-order nonlinear optical susceptibility inherent to a material that give rise to four wave mixing, and therefore provides a signal enhancement proportional to the cube of incident laser beam intensity, and also provides a mechanism of obtaining a three-dimensional image of an inner portion of the sample with a high spatial resolution, as in laser confocal microscopes. Moreover, since the CARS phenomenon is an optical parametric conversion process which does not dissipate any laser energy in the measured sample after the interaction, it is a non-invasive measuring method which can avoid thermal damage to the sample by the laser.
However, the CARS microscope has also some disadvantages that the molecular selectivity and the signal-to-noise ratio are lowered by the non-resonant third-order nonlinear optical susceptibility which is not relevant to the natural vibration of molecules, and the measurement sensitivity is still low in comparison with a method using direct resonance absorption in the mid-IR range.